Deploy: b36f4e2d5ac2bb930c0a8bf385fedd7dc2d04968

README.md

@@ -16,28 +16,8 @@ This backend runs a dual-model pipeline for text and image understanding.
 
 ## 🛠️ Supabase SQL Setup
 
-Run this in your Supabase SQL Editor:
-
-```sql
--- 1. Create Storage Bucket for User Uploads
-insert into storage.buckets (id, name, public) 
-values ('user_uploads', 'user_uploads', true);
-
--- 2. Create MLOps Training Logs Table
-create table training_logs (
-  id bigint generated by default as identity primary key,
-  created_at timestamp with time zone default timezone('utc'::text, now()) not null,
-  original_text text,
-  enhanced_text text,
-  vision_description text,
-  image_path text,
-  category text
-);
-
--- 3. Set up Storage Policies (Allow Public Read)
-create policy "Public Access" on storage.objects for select using ( bucket_id = 'user_uploads' );
-create policy "Allow Upload" on storage.objects for insert with check ( bucket_id = 'user_uploads' );
-```
+Canonical SQL is maintained in `docs/setup/supabase-schema.md`. Use that file
+as the source of truth and keep it in sync with any schema or storage changes.
 
 ## 🚀 API Deployment